Overall Aim: evaluate effectiveness of a selective, comprehensive, protocol-driven, school-based summer program designed to reduce risk & strengthen protective factors related to drug use/problem behaviors among youth. Intervention addresses limitations of school-based programs. Intervention is delivered m-F for 6 weeks in the summer between 4th & 5th grade (2 weeks academic readiness, 2 weeks residential, outdoor leadership camp, 2 more weeks academic readiness). Academic underachievement (80 hours); camp experience to foster resiliency & reduce peer rejection (160 hours). 5th grade teachers & high school mentors deliver intervention. This prospective, longitudinal, cohort-sequential design study includes all 4th grade students in 4 elementary schools in Louisville, KY (about 90 4th grade students attend each school). All 4th graders in each school will be ranked by principal/counselor on continua of academic & behavioral problems: 30 of 60 most problematic Ss will be randomly selected as intervention Ss, 30 not selected will be control Ss, the remaining 30 comparison Ss. The first cohort will be followed to 8th grade & so on in cohort sequential design. The total n will be 1,440; 480 intervention, control, & comparison Ss each. Three 5th grade teachers will deliver intervention in each school; 1 from the students' school. Intervention, control, & comparison students will be randomly assigned to 5th grade classes. Unit of analysis will usually be individual student (design allows for nesting Ss in classrooms, schools, communities. Covariance structure analysis & HLM 2/3 will be used to test measurement model which is reflected in the SPECIFIC AIMS. 1: measure incidence, prevalence, overlap, & correlates of child problem behaviors of attention deficit hyperactivity disorder (ADHD), attention deficit deficit disorder (ADD), conduct disorder (CD), & oppositional defiant disorder (ODD); 2: examine role of child problem behaviors (ADHD, ADD, CD, ODD in predicting peer rejection & academic underachievement, peer influence, & various forms of prosocial & antisocial behavior; 3: examine role of intrapersonal traits (novelty seeking, harm avoidance, reward dependence, persistence) in predicting child behavior problems & in mediating effects of family & neighborhood influence on child problem behaviors; 4: examine role of family influences in predicting intrapersonal traits child problem behaviors; 5: examine role of neighborhood influences in predicting intrapersonal traits & child problem behaviors; 6: examine role of peer influence in mediating the effects of peer rejection & academic underachievement, child problem behaviors, & family & neighborhood influence on adolescent prosocial & antisocial behavior; 7: test model within neighborhood contexts; 8: test model sorting sample into early/late starters, life-course-persistent/adolescence-limited pathways to delinquency.